Positive surgical margins after mini-invasive radical prostatectomy: a multi-institutional study

POSITIVE SURGICAL MARGINS AFTER MINI-INVASIVE RADICAL PROSTATECTOMY: A MULTI-INSTITUTIONAL STUDY A. Tafa, G. Albo, E. De Lorenzis, D. Consonni, A. Porreca, A. Celia, A. Antonelli, M. Falsaperla, A. Minervini, P. Parma, S. Zaramella, P. Bove, C. Ceruti, S. Crivellaro, B. Rocco (Milano) Aim of the study Positive surgical margins (PSMs) are a known risk factor for biochemical recurrence in prostate cancer (PCa) and are potentially affected by surgical technique. We investigated whether mini-invasive radical prostatectomy (RP) modality affects PSM rates and analyzed the incidence and the associative factors for PSMs in a multi-institutional series of 1357 robotic and 635 laparoscopic RP. Materials and methods We retrospectively analyzed 1992 patients with clinically localized PCa treated with laparoscopic RP (LRP) or robot assisted RP (RARP) by one of 35 surgeons from eleven institutions in Italy between January 2008 and March 2014. We had no data about margin status in 177 patients (123 LRP, 54 RARP). Patients were excluded from analysis if they had previously received androgen deprivation therapy or radiation therapy to the pelvis. PSMs were defined as cancer at the inked margin. 1159 patients were available for multivariate analysis. Results PSM rates were 24.1% and 20.2% for RARP and LRP respectively (p=0.074). Pathological stages were pT2 in 84,6% and pT3 in 15,4 % of patients for LRP and pT2 in 75,8% and pT3 in 24,2% of patients for RARP (p Discussion In our cohort PSA and NS procedure had no statistically significant impact on the PSM rates on multivariate analysis. There was no significant difference on PSM rates between the two analyzed surgical technique, however RARP tended to have higher rates. This may be explained with more pT3 and pGS ≥ 7 patients present in the RARP cohort, which were the most important predictive factors for PSMs on multivariate analyses, and the possible presence of the robotic RP surgeons at their initial learning curve. Limits of this study are: it’s not randomized nature, missing data across covariates, lack of central pathology review, lack of information for potential confounders (comorbidity, tumor volume, surgeon caseload and non entering on multivariate analysis other variables such body mass index, preoperative GS and clinical stage). Conclusions There is no significant difference between LRP and RARP for PSMs, with tendency of the robotic modality to operate higher pT stage PCa patients. Pathological stage and postoperative GS were the most important factors independently associated with an increased risk of PSMs after mini-invasive RP.